Digital analysis of discrete fractional order worms transmission in wireless sensor systems: performance validation by artificial intelligence

This article deals with a novel non-linear discrete fractional-order mathematical model connected with the spread of worms in the wireless sensor systems (WSSs). The proposed model classified into five classes such as, susceptible individuals, exposed individuals, infectious individuals, recovered individuals, vaccinated individuals (software installation). This model provides a complete framework for insight the spread of viruses in vulnerable systems and recommends potential countermeasures. This study shows that the mean squared error (MSE) in the testing phase is minimized, signifying accurate predictions. Levenberg-Marquardt neural network analysis and artificial intelligence technique have been utilized to estimate the model’s performance, incorporating its training status, regression analysis, error distribution, and overall suitability. The model is fractionalized via discrete Caputo operator, while the existence and uniqueness of results are obtained through fixed-point theory. Numerical simulations demonstrate the model’s usefulness in capturing the transmission dynamics of malicious codes. The model data has been divided into specific proportions: 70% for training, 15% for validation, and 15% for testing. Numerical results are achieved to support and justify the results for different fractional order.